This study utilizes Deep Neural Networks (DNN) to improve the K-Profile Parameterization (KPP) for the vertical mixing effects in the ocean’s surface boundary layer turbulence. The DNNs were trained using 11-year turbulence-resolving solutions, obtained by running a large eddy simulation model for Ocean Station Papa, to predict the turbulence velocity scale coefficient and unresolved shear coefficient in the KPP. The DNN-augmented KPP schemes (KPP_DNN) have been implemented in the General Ocean Turbulence Model (GOTM). This implementation is stable for long-term integration and as efficient as existing variants of KPP schemes. Three different KPP_DNN schemes, varying in input and output variables, have been developed and trained. The performance of models using the KPP_DNN schemes is compared with that of those using popular deterministic first-order and second-moment closure turbulent mixing parameterizations. Solution comparisons show that the simulated mixed layer is cooler and deeper, aligning closely with observations when wave effects are included in parameterizations. In the KPP framework, changes to the velocity scale of unresolved shear, which is used to calculate mixed layer depth, have a larger impact on the simulated mixed layer than do changes to the magnitude of diffusivity. In the KPP_DNN, changes to unresolved shear depend on not only on wave forcing, but also on the mixed layer depth and buoyancy forcing.

Wind, wave, and acoustic observations are used to test a scaling for ambient sound levels in the ocean that is based on the relative penetration depth of active bubbles during surface wave breaking. The focus is on acoustic frequencies in the range 1-10 kHz, which are typically scaled by wind speed alone. Wind and wave information are combined in a parametric form to describe the depth of the active bubble layer (which produces sound) relative to the depth of the passive bubble layer (which attenuates sound). The relative depth scaling has a primary dependence on wind speed and a secondary dependence on any departure of significant wave height from fully-developed, open-ocean conditions. The scaling is tested with long time-series observations of winds and waves at Ocean Station Papa (North Pacific Ocean), as well as with a case study with fetch limitation near the island of Jan Mayen (Norwegian Sea). When waves are less developed (e.g., limited by fetch) at a given wind speed, the attenuating layer is relatively thin and the sound levels are higher. The scaling is a plausible explanation for the observed reduction in sound levels during high wind events (winds greater than 15 m/s).

In the past decade, two large marine heatwaves (MHWs) formed in the northeast Pacific near Ocean Station Papa (OSP), one of the oldest oceanic time series stations. Physical, biogeochemical and biological parameters observed at OSP from 2013 to 2020 are used to assess ocean response and potential impacts on marine life from the 2019 northeast Pacific MHW. The 2019 MHW was preceded by calm and stratified surface conditions, lower dissolved inorganic carbon, and higher pH of surface waters relative to the 2013-2020 period. A spike in the summertime chlorophyll followed by a decrease in surface macronutrients suggests increased productivity in the well-lit stratified upper ocean during summer 2019. More blue whale calls were recorded at OSP in 2019 compared to the prior year. Large subsurface temperature anomalies were also found, suggesting that the earlier northeast Pacific MHW (2013-2015, previously referred to as “Blob”) as well as the long-term increase in sea surface temperatures in the region contributed to the intensity of the 2019 MHW. This study shows how the utility of long-term, continuous oceanographic datasets and analysis with an interdisciplinary lens is necessary to understand the potential impact of MHWs on marine ecosystems.